Feeler: EZ Tube Lunch Box

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
dipfrik said:
If i'am compare this solution with the one from Holgers initial lunchbox, there is a huge difference in the required size to be capable of providing the power rails to the motherboard or modules directly. What kind of connector and more important cable/wire size you think are save in our use case? Which size (AWG) of cable can the output connector pins carrying?

Cheers, Sven

The two key parameters we need to consider the maximum voltage and the maximum current. The HT is the maximum voltage at about 300 volts. The connector I have chosen is rated at 600V so there is plenty of margin. The heaters draw the most current. The power supply is designed to power up to 6 modules and as each one consumes 0.45 amps of heater current, the maximum heater current will be just under 3 amps. The contacts in the connector I have chosen are rated at 7.5 amps so once again there is plenty of margin.

There are several alternative pins that can be used in the connector. One type will take wire from 24 to 20AWG. A 20AWG wire has a resistance of 0.033 ohms per metre so if we have a 1m lead from an external supply, the total resistance will be 0.066 ohms. The voltage drop in this cable when carrying 3 amps will be 0.099 volts i.e. less than 0.1V which is satisfactory.

For the HT we can use 24 AWG wire which has nearly three times the resistance. However, the HT current is more than 10 times less than the heater current so the voltage drop will be negligible.

Of course, when the power supply is internal, the lead length is less than half a metre so the voltage drops will be even less.

Cheers

Ian
 
Hey Ian,

this would be so cool, if we have a connector which not feels like dealing with "City Power Current" ;)
Where do you find this connector, and is a suitable cable type somewhere available? (cable in terms of multiple wires in one cover)

Cheers, Sven
 
dipfrik said:
Hey Ian,

this would be so cool, if we have a connector which not feels like dealing with "City Power Current" ;)
Where do you find this connector, and is a suitable cable type somewhere available? (cable in terms of multiple wires in one cover)

Cheers, Sven

'This is the connector I am thinking of:

http://www.rapidonline.com/cables-connectors/cpc-9-pin-standard-pin-casing-with-rectangular-flange-50-0135

Inside the lunch box I would just use individual wires. For an external supply I have some 8 way screened cable I use but unfortunately you have to buy a reel of it. I use the 16/0.2 type which is close to 20AWG:

http://www.rapidonline.com/cables-connectors/unistrand-screened-signal-multicore-cable-defence-standard-82987

Cheers

Ian
 
Hey Ian,

ok, i will do some search here in germany for similar specs or order a few of these connectors plus wire from rapid and give it a try.

Cheers, Sven
 
dipfrik said:
Hey Ian,

ok, i will do some search here in germany for similar specs or order a few of these connectors plus wire from rapid and give it a try.

Cheers, Sven

There is a confusing range to choose from. They are all TE Connectivity (AMP) types. Here is a selection of 9 way types from Farnell:

http://uk.farnell.com/webapp/wcs/stores/servlet/Search?catalogId=15001&langId=44&storeId=10151&categoryId=700000005038&pageSize=25&beginIndex=1&showResults=true&pf=110171155,110179367,111644429,111644487

Cheers

Ian
 
Hey Ian,

short question beacuse I'am running into big problems with my big tube amp prototype pcb.

If a signal is only going to the 1st amp stage and from OUT1 directly to the output transformer, then there should be no signal in any case measurable at OUT2 or? Also if i'am feeding only a signal over IN2 to the 2nd amp stage and from there over OUT2 into the output transformer, there should be nothing at OUT1 measurable or?

Bottom line both amp stages are only connectable through IN2 in the case of an EQ signal coming in.
Or in other words, both amp stages are totally seperated by default.

I'am asking this, and hopefully the only answer is: "yes, OUT2 should be dead silent if you only using the 1st amp stage and no signal is visible at IN2" ;) , because my board are acting very strange at high gain settings with an EQ and the 2nd gain stage in use. Only using the 1st amp stage and everything seems fine.

After hours of searching, it was an accident, that i connected the OUT2 to the output transformer, which made no sense at this time, because there was no eq or anything connected to IN2, but i was able to pick up a  signal like going to an extreme high pass filter at OUT2.  Again nothing was connected to IN2 or anything.

Update: same happens with a signal going directly ONLY into IN2. I can hear something at OUT1.

So i bet this is a major bug here with my board, beacuse i'am expect a dead silent 2nd amp stage right?

Thanks, Sven
 
Yes, the amp not being used should be dead silent. However, the input impedance of each amp is very high so if IN2 for example is left unconnected it might be able to pick up some signal from the other amp.

Cheers

Ian
 
I have just finished building the new four toggles PCB before fitting it to the 'Classic' and testing it. The picture shows on the left the bare PCB which contains two sets of four toggles which you can use if you want to squeeze two mic channels into a single 24HP module. The one the right shows one half of the board after I sawed it in half with a junior hack saw and in the middle you can see the one I populated. The top switch is the 48V phantom power, the next is the 20dB pad for the mic input, the third is the mic/line switch and the bottom one is the phase switch that affects both the mic and line inputs. For this build, as I intend to use it with the Classic PCB I am building, I have used 10K resistors and a 150R resistor in the line input pad which gives an attenuation of 42.6dB. The gain of the first stage is 20dB (transformer) + 30dB (mu follower) = 50dB so a 0dBu input to the line input appears at +7.4dBu after the first stage.

For this build I put the resistors on the same side as the switch and the Molex connectors on the back of the PCB. I think the next time I will put the switches on the back and the resistors and Molex connectors on the front. That way when you fit the switches to the panel, the components and silk screen are  visible from the rear of the panel where you will be wiring up the PCB.

Cheers

Ian
 

Attachments

  • 4togglesPCBs.png
    4togglesPCBs.png
    738.4 KB · Views: 108
Hey Ian,

I have used 10K resistors and a 150R resistor in the line input pad which gives an attenuation of 42.6dB.

How can i calculate my own pad settings? There is some math behind?

Thanks, Sven
 
dipfrik said:
Hey Ian,

I have used 10K resistors and a 150R resistor in the line input pad which gives an attenuation of 42.6dB.

How can i calculate my own pad settings? There is some math behind?

Thanks, Sven

Good old Mr. Ohm is in charge of the maths for pads. Rough guide is as follows:

1. Assume we want to look like 150 ohms or thereabouts to the transformer - this means the output arm of our pot divider will be 150 ohms.
2. Decide how much attenuation we want and convert it to a loss. Let's say we want 20dB which is a loss of 10 times.
3. This means the total resistance of our pad must be 10 times 150 ohms or 1500 ohms.
4. As we already have 150 ohms, the other two resistors must total 1500 - 150 ohms = 1350 ohms.
5. For a balanced pad we have two resistors so each one is 1350/2 = 675
6. Choose the nearest standard value resistors i.e. 680 ohms and we are done.

Cheers

Ian
 
Some pics of the 'Classic' build. First the prototype front panel including the four toggles and a single level control:

Cheers

Ian
 

Attachments

  • classicfrontpanel.png
    classicfrontpanel.png
    865.1 KB · Views: 94
After a very unsuccessful week with my ez-tube-amp-prototype-monster-board, i started to get my hands dirty on 3U modules came in from buerklin. After i nearly crashed the S-AOC PSU, seeing resistors burned like hell (ok, this was a stupid idea i will no one tell ;)) and getting to know that mechanical work without the proper tools are useless, i finished fitting the psu and compressor into the modules and run my lovely tempest through. We will be friends forever ;)

I think i need a ton of literature about ("the concept") of grounding, shielding and at least placing components into the right position.

Maybe i will start with some new pcb design this weekend, because nothing really dangerous can happen there! ;)

Night, Sven

1st_modules.jpg
 
Apparently there is a Chinese curse which says 'May you live in interesting times'. Sounds like you had a very interesting few days. I am sure Holger will tell you that there is a lot more to designing and building mixers than just laying out PCBs!!

Cheers

Ian
 
ruffrecords said:
Apparently there is a Chinese curse which says 'May you live in interesting times'. Sounds like you had a very interesting few days. I am sure Holger will tell you that there is a lot more to designing and building mixers than just laying out PCBs!!

Cheers

Ian

Oh yes, he is and always will be! ;)

Cheers, Sven
 
Yesterday I finished the basic wiring and testing of the 3U PSU. I found a couple silly errors on the PCB. On both the 12.6V and 48V outputs, the output pads are connected to nothing!! I wondered why both outputs measured 0V. I have a simple workaround but I will obviously need to update the PCB layout before making it generally available. Just need to sort out the 9 pin connector and get hold of some different coloured 20AWG wire and I can complete it.

Cheers

Ian
 
Hey Ian,

two short question regarding the "EasyEQ" Integration.

1) is the integration on the attached screenshot correct?
2) are the pin/lug assignments of the pots correct and Pin 1 of the pots are not connected at all?

Thanks, Sven

EASYEQ_INTEGRATION.png
 
Hi Sven,

Yes, that is correct. On the current Twin Line Amp (TLA) PCB this is a lttle awkward to do because the electrolytic capacitor is connected to ground. The V2 version of the TLA has the gain pot and capacitor swapped and R3 is already fitted so it will be a lot easier.

Cheers

Ian
 

Latest posts

Back
Top